The goal of this project is to determine whether the appearance of epidermal blisters (acantholysis) characteristic of the autoimmune disease pemphigus foliaceus (PG) is caused by autoantibodies against the desmosomal glycoprotein desmoglein I (DG I). The strong possibility that this is so has been raised by the discovery that antibodies against this protein were uniquely present in all PF sera examined but not in other sera, including those from patients with pemphigus vulgaris. The evidence that DG I is "the pemphigus foliaceus antigen" is at this point purely correlative. In the experiments proposed here, experimental test of this hypothesis will be performed. This will be done by purifying DG I, using it to absorb from PF IgG specifically those antibodies directed against DG I, and separately testing, both in vitro (human foreskin) and in vivo (mouse), the acantholysis- inducing ability of the affinity-purified monospecific autoantibody and of the residual, specifically depleted autoimmune serum. It is hoped in this way to obtain a definitive functional test of the hypothesis that PF is brought about by an autoimmune attack upon DG I. Purification of minimally denatured bovine muzzle epidermis DG I will be attempted in several ways. Purification for detergent extracts of skin slices will be attempted by immunoprecipitation with PF IgG-derived or other antibodies against DG I. An attempt will also be made to remove the external domains of DG II from desmosomal core vesicles by controlled proteolysis, leaving only DG I external domains immunologically recognizable by externally applied PF IgG. Probably, however, we will need to purify DG I from desmosomal core vesicles by finding a way to solubilize it without resorting to the use of SDS. Possible procedures are suggested. Purified bovine epidermal DG I will be used as an affinity reagent to separate human anti-DG I IgG from PF whole IgG, and the ability of the separated fractions to induce acantholysis will be tested on fragments of human foreskins in vitro and in the in vivo mouse passive transfer assay. The reported effects of plasminogen and of protease inhibitors upon the results of the in vitro assay will also be examined.